Groundwater Recharge Dynamics in a Semi‐Arid Watershed Under a Changing Climate

Abstract

Groundwater is a critically important resource in the semi‐arid regions of Telangana, where its availability is highly sensitive to climatic variability and land use and land cover changes. This study investigates climate‐driven groundwater recharge dynamics in a semi‐arid hard‐rock watershed through comparative analysis of SWAT‐simulated groundwater recharge and long‐term groundwater‐level observations. The Soil and Water Assessment Tool (SWAT) was used to simulate hydrological processes and the model was calibrated and validated using a calibration and uncertainty approach. The model showed satisfactory performance with Nash–Sutcliffe efficiency (NSE) values of 0.54–0.57, coefficient of determination ( R 2 ) values of 0.61–0.66 and Percent Bias (PBIAS) within ±15%. Model‐derived recharge outputs were compared with observed piezometer measurements using normalised variability and lag correlation analysis (2002–2020), revealing a distinct 1‐year delayed aquifer response to annual recharge inputs. Future hydroclimatic impacts (2026–2100) were assessed using a multi‐model ensemble of five climate model projections from the Coupled Model Intercomparison Project Phase 6 (CMIP6) under four Shared Socioeconomic Pathways (SSP1‐2.6, SSP2‐4.5, SSP3‐7.0, SSP5‐8.5). Climate model evaluation showed good agreement with observations, with root mean square error (RMSE) and mean absolute error (MAE) values indicating acceptable performance across precipitation and temperature variables. Results indicate strong sensitivity of groundwater recharge to precipitation variability, with higher recharge expected under mid‐to‐high emission scenarios, along with increased evapotranspiration due to rising temperatures. Historical average groundwater recharge (1956–2014) was 209.44 mm, increasing to 220.31 mm under present climate conditions and is projected to reach 251.81 mm in future scenarios, representing an overall increase of about 20.22%, with greater fluctuations under high‐emission scenarios. The results further show that future recharge is expected to become increasingly variable over time, with high recharge occurring during specific periods under future climate conditions. These findings improve understanding of advances in hydrological process by explicitly linking SWAT‐simulated recharge to observed piezometer responses, quantifying a one‐year aquifer transmission time and assessing how this transmission and recharge magnitude change under CMIP6 scenarios.